Since soil properties influence the behavior of soils, the knowledge related to these properties is important in using them for different agricultural purposes. This study aimed to develop a structural equation model of yield components of wheat (YCW) in northwest of Iran using soil physical and chemical properties. Soil samples were collected from Mollaahmad watershed of Ardabil province in northwest of Iran for the greenhouse experiment. The primary purpose of this research was to develop a conceptual model in order to determine the sources of variations within the dataset and to explore equations for the sampled soils. The findings revealed that two soil properties components (chemical and physical properties) were significant in explaining YCW. The accepted model in the multiple linear regression (MLR) analysis demonstrated that the soil's chemical and physical properties measures are statistically significant in estimating YCW. Following this, and according to R square statistic, 87% of the variance in YCW was explained by the soil chemical properties and 83% was accounted for by soil physical properties. Considering the relative importance of the estimation of YCW variable and from the perspective of regression equations, the organic carbon and saturated point moisture made the largest contribution through the two proposed models for the soil productivity. According to the structural equation modeling (SEM) results, the final model has proved that YCW was controlled by soil chemical properties more than physical properties. The obtained general model can be useful for wheat and also an analytical pattern for Gramineae family. The improved estimation of production might be valuable in practice because crop productions are widely applied, for instance, to assess agroenvironmental policy measures to compare cropping systems or the need of a better soil quality managing.
